The MEK5 [MAPK (mitogen-activated protein kinase)/ERK (extracellular-signal-regulated kinase) kinase 5]/ERK5 pathway is the least well studied MAPK signalling module. It has been proposed to play a role in the pathology of cancer. In the present paper, we review the role of the MEK5/ERK5 pathway using the ‘hallmarks of cancer’ as a framework and consider how this pathway is deregulated. As well as playing a key role in endothelial cell survival and tubular morphogenesis during tumour neovascularization, ERK5 is also emerging as a regulator of tumour cell invasion and migration. Several oncogenes can stimulate ERK5 activity, and protein levels are increased by a novel amplification at chromosome locus 17p11 and by down-regulation of the microRNAs miR-143 and miR-145. Together, these finding underscore the case for further investigation into understanding the role of ERK5 in cancer.
ERK (extracellular-signal-regulated kinase) 5 is the effector kinase of a canonical three-tiered MAPK (mitogen-activated protein kinase) signalling cascade comprising MEK (MAPK/ERK kinase) 5, MEKK (MEK kinase) 2/3 and ERK5 itself [1,2]. The ERK5 protein, encoded by the MAPK7 gene , contains an N-terminal kinase domain that shares 50% identity with ERK2, and a large C-terminal extension that contains a transactivation domain, an NLS (nuclear localization sequence), an NES (nuclear export sequence) and two proline-rich regions [1,3,4] (Figure 1). Owing to its large size, ERK5 is sometimes referred to as BMK1 (big MAPK1) .
Schematic representation of the role and regulation of ERK5 by growth factors, oncogenes and during mitosis
In normal physiology, MEK5 and ERK5 are ubiquitously expressed [1,2,5] and are activated by growth factors and cellular stresses [6–8]. In development, the MEK5/ERK5 pathway is important for blood vessel and cardiac development, as removal of Mek5 or Erk5 is embryonic lethal at E (embryonic day) 9.5–10.5 with defects in these tissues [9,10]. Using an in vitro muscle differentiation system, ERK5 has also been shown to be important for muscle development , whereas, in the adult, the MEK5/ERK5 pathway is important in regulating the proliferation and survival of endothelial cells [12–14] and cells of the immune system [15,16].
In cancer, there is clinical evidence that an increase in MEK5/ERK5 signalling may be important for disease progression. For example, in breast cancer, MEK5 expression is up-regulated by constitutive activation of STAT (signal transducer and activator of transcription) 3 . Constitutive activation of STAT3 is frequently detected in patients with advanced breast cancer, but not in normal breast epithelial cells . Furthermore, increased ERK5 protein levels are associated with decreased disease-free survival . In prostate cancer, MEK5 is overexpressed and correlates with the presence of bone metastases and less favourable disease-specific survival . Furthermore, ERK5 is a target for gene amplification at 17p11 in HCC (hepatocellular carcinoma), an amplification detected in approximately 50% of primary HCC tumours .
ERK5 activity has also been shown to be activated by some oncogenes, and its protein levels are subject to regulation by tumour-suppressive miRNAs (microRNAs), suggesting that targeted therapies against ERK5 may have a more widespread clinical application than the cancers described above. In the present review, we assess the rationale for targeting ERK5 using the ‘hallmarks of cancer’  as a framework, discuss the evidence for ERK5 activation by oncogenes and consider the mechanisms that regulate ERK5 protein levels.
ERK5 and the hallmarks of cancer
The hallmarks of cancer are an organizing framework of biological capabilities that allows the complexity of tumour development to be rationalized. The eight hallmarks are sustaining proliferative signalling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, activating invasion and metastasis, reprogramming of energy metabolism, and evading immune destruction. These hallmarks are acquired by normal cells during tumour pathogenesis to produce tumorigenic and malignant cells . However, it is important to note that tumours are not simply an isolated mass of proliferating cancer cells; the tumour microenvironment aids tumorigenesis as there are essential interactions between the surrounding stromal cells and the tumour cells . Nevertheless, the hallmarks of cancer provide a critical framework for assessing the role of genes in promoting or sustaining the cancer cell phenotype.
Sustaining proliferative signalling
Cancer cells sustain proliferation by autocrine proliferative signals, overexpression of cell-surface receptors and/or constitutive activation of signalling pathway components. They typically become ‘addicted’ to key pathways for their proliferation (and survival), therefore defining the components of these pathways is important. ERK5 was first demonstrated to be important for cancer cell proliferation in HeLa cells , where a dominant-negative form of ERK5 (the residues phosphorylated by MEK5 are mutated to alanine or phenylalanine: an AEF motif rather than a TEY motif) blocked entry into S-phase in response to EGF (epidermal growth factor). Similar studies have suggested a role for ERK5 in the proliferation of the MCF7 and BT474 breast cancer cell lines , the immortalized breast epithelial cell line MCF10A  and in multiple myeloma MM1S cells . Furthermore, expression of a constitutively active form of MEK5 (the activation-loop residues are mutated to aspartic acid to mimic phosphorylation; MEK5D) increased the proliferation of HEK (human embryonic kidney)-293 cells and the prostate cancer cell line LNCaP . RNAi (RNA interference) techniques have confirmed the requirement for ERK5 in cell proliferation of LNCaP and C4-2 cells  and in T24 bladder cells . It should be noted that, although ectopic expression of ERK5 in the prostate cancer cell line PC3 increased proliferation, RNAi knockdown did not decrease proliferation . This suggests that the role of ERK5 in sustaining the proliferative signal may be cancer-cell-line- or tumour-specific, and at this time, it is unknown what makes a cancer cell sensitive to ERK5 inhibition.
Resisting cell death
Although cancer cells are under continual stress caused by oncogenic signalling and DNA damage associated with hyperproliferation, they have devised mechanisms to evade apoptosis. The most common mechanism is loss or inactivation of the tumour suppressor p53. Other mechanisms include up-regulation of pro-survival proteins (i.e. Bcl-2) or down-regulation of pro-apoptotic proteins (i.e. Bax and Bim). These proteins are under direct regulation by signalling cascades. The MEK5/ERK5 cascade has been shown to protect cells from apoptotic stimuli as Mek5−/− MEFs (mouse embryonic fibroblasts) are more sensitive to osmotic-stress-induced apoptosis  and overexpression of a nuclear-localized ERK5 mutant (Δ570) protects HeLa cells from TRAIL [TNFα (tumour necrosis factor α)-related apoptosis-inducing ligand]-induced apoptosis . Indeed, MEK5 and ERK5 are required for resistance of MCF7 cells to apoptosis induced by etoposide, TRAIL and TNFα . Overexpression of ERK5 protects multiple myeloma cells from PS431-induced apoptosis, whereas overexpression of the non-activatable mutant (AEF) sensitizes them to apoptosis induced by PS431 and dexamethasone . ERK5 knockdown by shRNA (small hairpin RNA) sensitizes T-lymphocytes to Fas-induced apoptosis. This pro-survival role of ERK5 was mediated by NF-κB (nuclear factor κB) activation and nuclear localization . NF-κB nuclear localization has been observed in many cancers  and is thought to play an important role in promoting cell survival. Despite these examples, there is relatively little direct evidence that cancer cells have evolved to be dependent upon endogenous ERK5 signalling for survival, certainly in comparison with recognized cell survival kinases such as PKB (protein kinase B) and ERK1/2.
Activating invasion and metastasis
Cancers of epithelial origin progress to invade local tissues and metastasize to distal sites. This requires changes in cell shape, decreased attachment to other cells and the ECM (extracellular matrix) and an increase in cell motility. In advanced prostate cancer, high MEK5 levels are associated with bone metastasis  and in advanced OSCC (oral squamous cell carcinoma), high ERK5 expression is associated with the presence of lymph node metastases . In keeping with these observations, overexpression of ERK5 in an orthotopic model of prostate cancer induced metastasis to the lymph nodes and lungs . In experimental systems, ERK5 has been demonstrated to have a role in the formation of invadopodia in A375 malignant melanoma and PC3 prostate cancer cells  and podosomes in MEFs . Invadopodia and podosomes are dynamic, actin-based protrusions of the plasma membrane that are involved in cell attachment to the ECM. They also contain MMPs (matrix metalloproteases) that degrade the ECM and facilitate cell motility/invasion . Consistent with this function, ERK5 regulates the expression of MMP2 and MMP9 [20,27] and the degradation of synthetic ECM in vitro [27,33]. ERK5 also plays a role in regulating actin dynamics and cell motility. For example, ERK5 expression leads to the loss of stress fibres in NIH 3T3 cells, and inhibition of both ERK5 and ERK1/2 was required to restore stress fibres in Src-transformed NIH 3T3 cells . Furthermore, reduction of ERK5 levels by siRNA (small interfering RNA) reduced both random cell migration and invasion into Matrigel in PC3 cells  and HGF (hepatocyte growth factor)-induced cell migration of MDA-MB-231 breast cancer cells . This latter observation has been demonstrated to involve the RNA processing protein, SAM-68 (Src-associated in mitosis 68 kDa) . Taken together, these observations show that, in cancer cells with aberrant ERK5 activity, ERK5 may play a key role in co-ordinating migration and invasion.
Tumours require a blood supply to deliver nutrients and oxygen and to remove metabolic waste and carbon dioxide. Angiogenesis is the sprouting of new blood vessels from existing ones and, in tumour progression, there is an ‘angiogenic switch’ that causes quiescent vasculature to continually sprout new blood vessels to sustain the growing tumour. VEGF (vascular endothelial growth factor) is a well-known inducer of angiogenesis. Genetic deletion of Mek5 and Erk5 has demonstrated their important role in blood vessel formation [9–11]. In addition, the role of ERK5 signalling in tumour-associated neovascularization has also been tested directly. Targeted deletion of ERK5 in endothelial cells reduced the mass and vascular density of tumours in two xenograft models (B16F10 melanoma and LL/2 Lewis lung). Similar results were seen in Matrigel plug assays measuring neovascularization in response to VEGF and bFGF (basic fibroblast growth factor). It is thought the role that ERK5 plays in neovascularization is by mediating endothelial cell proliferation and survival . However, in an in vitro angiogenesis system, ERK5 is required for VEGF-induced tubular morphogenesis, but not cell proliferation. During tubular morphogenesis, ERK5 is required for VEGF-stimulated PKB phosphorylation on Ser473 and Thr308, Bad phosphorylation on Ser136 and inhibition of caspase 3/7 activity . These findings demonstrate the ERK5 is a pivotal enzyme in regulating tumour angiogenesis.
Regulation and deregulation of ERK5
Model of ERK5 activation and regulation
In order to understand whether ERK5 is deregulated in cancer, we must first consider how ERK5 is regulated under ‘normal’ conditions and how it mediates its functions. The canonical activation mechanism of ERK5 involves MEK5 phosphorylation of the TEY motif in the activation loop of the ERK5 kinase domain followed by autophosphorylation of the C-terminal extension on multiple sites [5,39] (Figure 1). The C-terminal extension inhibits the ERK5 kinase domain, whereas C-terminal autophosphorylation relieves this autoinhibition . The C-terminal extension of ERK5 is also phosphorylated by a MEK5-independent mechanism during mitosis, which requires the activity of CDK1 (cyclin-dependent kinase 1) (although ERK5 may not be a direct CDK1 substrate) [40–42] (Figure 1). Once active, and phosphorylated, ERK5 can function as a kinase to phosphorylate substrates, but can also function as a transcriptional transactivating factor [3,43] and play a role in transcriptional elongation . ERK5 is present in both the nucleus and the cytosol and probably exists in both inactive and active states in both compartments. However, MEK5 phosphorylation of the ERK5 activation loop and C-terminal autophosphorylation leads to an increase in the nuclear localization of ERK5. There is one non-synonymous somatic mutation in ERK5 reported in the Catalogue of Somatic Mutations in Cancer (COSMIC) (http://www.sanger.ac.uk/genetics/CGP/cosmic/). But with over 900 samples sequenced, this makes mutation of ERK5 a very infrequent event in cancer. Deregulation of the MEK5/ERK5 pathway is more likely to be achieved by sustained activation and increased protein levels.
Regulation of ERK5 by growth factors and oncogenes
Epidermal growth factor
EGF plays important role in the regulation of cell proliferation, survival and differentiation. EGF binds to the EGFR (EGF receptor), activates its intracellular tyrosine kinase, which in turn activates intracellular signalling cascades. EGF has been shown to activate ERK5 in many cell types [14,23,24,45–47]. NRG (neuregulin), proteins structurally related to EGF that activate erbB receptor tyrosine kinases (erbB2–erbB4) and activate ERK5 in MCF7 breast cancer cells . Activation of ERK5 in these systems is primarily involved in promoting cell proliferation, so it will be interesting to determine whether ERK5 is required in cancers that contain overexpressed or mutated EGFR.
Src, a proto-oncogenic non-receptor tyrosine kinase, is required for activation of the ERK5 pathway in response to some growth factors and cytokines. For example, Src is required for ERK5 activation by PDGF (platelet-derived growth factor) in hepatic stellate cells  and PAE (porcine aortic endothelial) cells , but not by the cytokine IL-6 (interleukin 6) in multiple myeloma cells . Src is also required for ERK5 activation by cellular stresses such as hydrogen peroxide in fibroblasts  and asbestos in lung epithelial cells , but not by fluid shear stress in endothelial cells . Oncogenic forms of Src (v-Src and SrcY527F) can activate the ERK5 pathway in fibroblasts [33,35,49], leading to an increase in ERK5 nuclear localization . Furthermore, ERK5 is required for Src-driven transformation in NIH 3T3 fibroblasts, as dominant-negative forms of MEK5 (where the residues required to be phosphorylated for activity are mutated to alanine; AA) and ERK5 (AEF) block foci formation . This raises the possibility that ERK5 is required to promote transformation in cancers where Src is deregulated.
Ras and Raf
Ras is a small GTPase that activates the Raf/MEK1/2/ERK1/2 pathway. There are three human Ras proteins, K-Ras, H-Ras and N-Ras, one or other of which is mutated in approximately 20% of all tumours. The Raf serine/threonine kinases (B-Raf, C-Raf and A-Raf) are activated directly by GTP-bound Ras proteins, and in turn activate the MEK1/2/ERK1/2 pathway. B-Raf is also mutated and thereby activated at high frequency in certain tumours. Studies indicate that the role of Ras in ERK5 activation is highly cell-type-specific; for example, a dominant-negative form of H-Ras (N17) inhibited ERK5 activation by EGF in rat phaeochromocytoma PC12 cells and mouse myoblast C2C12 cells, but not in COS7 cells . NRG-driven activation of ERK5 was independent of Ras in breast cancer cells  and IL-6 activation of ERK5 was also Ras-independent in multiple myeloma . Interestingly, an active form of H-Ras (12V) can activate the kinase domain of ERK5 in co-expression studies [23,51], although the physiological relevance of this is unclear. Similarly, MEK5D co-operates with C-Raf (Raf-1) to transform NIH 3T3 cells , raising the possibility that in cancers that harbour Ras and Raf mutations, ERK5 may be required for cell transformation. However, this hypothesis has not yet been thoroughly tested.
Tpl2 (tumour progression locus 2)/COT
COT is a serine/threonine protein kinase that has been implicated in cellular transformation. It is a member of the MAPK kinase kinase family and is overexpressed in human gastric/colon adenocarcinomas, large granular T-cell neoplasias, breast cancer and EBV (Epstein–Barr virus)-related nasopharyngeal carcinoma and Hodgkin's disease . COT causes cell transformation of NIH 3T3 fibroblasts by increasing c-Jun expression, and this is mediated by co-operative signals from ERK5, p38γ and JNK (c-Jun N-terminal kinase) .
Increased expression of ERK5 in cancer
Amplification of ERK5
Although activating mutations in ERK5 are unknown, a novel amplification has been identified at 17p11 in HCC (hepatocellular carcinoma). This contains the MAPK7 ERK5-encoding gene and ERK5 protein is highly overexpressed in HCC cells that exhibit this amplicon. This increase in MAPK7 copy number was detected in 35/66 primary HCC tumours. The SNU449 HCC cell line with the 17p11 amplification requires ERK5 for proliferation . There are four reports of other cell lines with amplifications on the Cancer Genome Project's Copy Number Analysis (CONAN) website (http://www.sanger.ac.uk/cgi-bin/genetics/CGP/conan/search.cgi). It will be interesting to determine whether these cell lines have also developed ERK5 ‘addiction’, similar to SNU449.
Regulation by miR-143
An exciting new level of regulation of ERK5 protein levels has recently been discovered, namely regulation by the tumour-suppressive miRNAs miR-143 and miR-145 [25,26,55,56]. miRNAs regulate mRNA function by modulating mRNA stability and translation. They are small non-coding single-stranded RNAs of approximately 22 nt in length and primarily function by binding to the 3′-UTR (untranslated region) of target mRNAs . miR-143 and miR-145 are both encoded by the same gene  and have both been shown to down-regulate ERK5 protein levels [25,26,55,56]. Interestingly, miR-143 expression is decreased in prostate cancer [25,27] and B-cell malignancies  and the bladder cancer cell line T24 , where ERK5 expression is increased. Reintroduction of synthetic miR-143 and miR-145 reduces ERK5 levels and decreases cell proliferation of T24 bladder cancer cells , LNCaP and C4-2 prostate cancer cells  and Burkitt's lymphoma Raji cells . Similar results were seen with RNAi knockdown of ERK5 in LNCaP and C4-2  and T24 cells , suggesting that ERK5 is responsible for these effects. Although miR-143 and miR-145 will have multiple targets within the cell, it is possible that, in cancers with down-regulated miR-143 and miR-145, ERK5 may play a role in the aetiology of the disease.
Small-molecule inhibitors of MEK5 and ERK5
There are currently three small-molecule inhibitors of the MEK5/ERK5 pathway. The first to be described were BIX02188 and BIX02189 by Boehringer Ingelheim Pharmaceuticals . These are selective inhibitors of MEK5 and do not inhibit MEK1/2. These have been used to demonstrate a role for the MEK5/ERK5 pathway in nerve-growth-factor-induced neurite outgrowth and stabilization of tyrosine hydroxylase in PC12 cells . However, to date, there are no data describing the effects of these inhibitors on cancer cell proliferation, survival or migration. The third newly described inhibitor is XMD8-92, a selective inhibitor of ERK5. Excitingly, XMD8-92 has anti-proliferative effects on HeLa cells and inhibited the growth of human tumour xenografts [HeLa (Nod/Scid)] and syngenic mouse xenografts [LL/2 (B6)], suggesting that ERK5 inhibitors may be potential cancer new therapies .
It is clear that the MEK5/ERK5 pathway plays an important role in tumour neovascularization, and there is an emer-ging role in tumour cell invasion and metastasis. Furthermore, there are multiple ways in which this pathway may be deregulated in cancer. This suggests that MEK5- or ERK5-targeted therapies may have a more widespread clinical application than first anticipated. With the advent of selective small-molecule inhibitors of MEK5 and ERK5, and novel miRNA therapeutic technologies, our understanding of the role and requirement of MEK5 and ERK5 in cancer will continue to grow.
Signalling 2011: a Biochemical Society Centenary Celebration: A Biochemical Society Focused Meeting held at the University of Edinburgh, U.K., 8–10 June 2011. Organized and Edited by Nicholas Brindle (Leicester, U.K.), Simon Cook (The Babraham Institute, U.K.), Jeff McIlhinney (Oxford, U.K.), Simon Morley (University of Sussex, U.K.), Sandip Patel (University College London, U.K.), Susan Pyne (University of Strathclyde, U.K.), Colin Taylor (Cambridge, U.K.), Alan Wallace (AstraZeneca, U.K.) and Stephen Yarwood (Glasgow, U.K.).
cyclin-dependent kinase 1
epidermal growth factor
mitogen-activated protein kinasel
mouse embryonic fibroblast
nuclear factor κB
protein kinase B
signal transducer and activator of transcription
tumour necrosis factor α, TRAIL, TNFα (tumour necrosis factor α)-related apoptosis-inducing ligand
vascular endothelial growth factor
We apologize to those authors whose work we have not been able to cite due to limited space in the present review.
Our work is supported by the Babraham Institute, the Biotechnology and Biological Sciences Research Council and grants from the Association for International Cancer Research and Cancer Research UK. P.A.L. is supported by a grant from the Association for International Cancer Research awarded to S.J.C.